Lyft, Inc.

Swarn Avinash Kumar

Joan Devassy

Michal Witkowski

Nadha Nafeeza Gafoor

Wei You

Ritwik Subir Das

John Michael Sparks

Angela Lam

Wiebke Poerschke

Hristo Stefanov Stefanov

Hugo Oscar Bonay Grimmett

Aahel Iyer

Mousom Dhar Gupta

Kia Nishimine

Examples disclosed herein involve a computing system configured to (i) obtain (a) a first set of sensor data captured by a first sensor system of a first vehicle that indicates the first vehicle's movement and location with a first degree of accuracy and (b) a second set of sensor data captured by a second sensor system of a second vehicle that indicates the second vehicle's movement and location with a second degree of accuracy that differs from the first degree of accuracy, (ii) based on the first set of sensor data, derive a first trajectory for the first vehicle that is defined in terms of a source-agnostic coordinate frame, (iii) based on the second set of sensor data, derive a second trajectory for the second vehicle that is defined in terms of the source-agnostic coordinate frame, and (iv) store the first and second trajectories in a database of source-agnostic trajectories.

Systems and methods for generating source-agnostic trajectories

This disclosure describes a card-scan system that can update a card-scan machine learning model to improve card-character predictions for payment cards, driver licenses, or other character-bearing cards by using an active-learning technique that learns from card-scan representations indicating corrections by users to predicted card characters. In particular, the disclosed systems can use a client device to capture and analyze a set of card images of a character-bearing card to predict card characters using a card-scan machine learning model. The disclosed systems can further receive card-scan gradients representing one or more corrections to incorrectly predicted card characters. Based on the card-scan gradients, the disclosed systems can generate active-learning metrics and retrain or update the card-scan machine learning model based on such active-learning metrics. By utilizing and updating the card-scan machine learning model, the disclosed systems can improve the accuracy with which card-character-detection systems predict card characters while preserving data security and verifying the presence of a physical character-bearing card.

Automatically updating a card-scan machine learning model based on predicting card characters

Growing community of inventors

Mountain View, CA, United States of America

Swarn Avinash Kumar